This article addresses limitations of Logistics Performance Index (LPI) and suggests remedies. Reliability of the instrument used in LPI may be better found by Angular Association method or Bhattacharyya’s measure, using only the frequencies or probabilities of item–response categories without involving assumptions of continuous nature or linearity or normality for the observed variables or the underlying variable being measured. The suggested methods also avoid test of uni-dimensionality, assumption of normality, bivariate normality. The problems of outlying observations and linear assumptions in principal component analysis for finding reliability theta are also avoided in each proposed method. Geometric mean approach provides a better alternative to compute LPI scores avoiding scaling and calculation of weights satisfies many desired properties and reduces level of substitutability between components, facilitates statistical test of equality of two geometric means and identifies critical areas for corrective measures. Such identifications are important from a policy point of view. The graph of LPI for a country over a long period of time reflects pattern of growth of LPI for the country. The method helps to rank and benchmark the countries, if the target vector is taken as LPI score of the best performing country. JEL Codes: C43, C54

中文翻译：

---

物流绩效指数：方法论问题

本文解决了物流绩效指数 (LPI) 的局限性并提出了补救措施。LPI 中使用的工具的可靠性可以通过 Angular Association 方法或 Bhattacharyya 的度量更好地找到，仅使用项目-响应类别的频率或概率，而不涉及对观察变量或被测量的基础变量的连续性或线性或正态性的假设. 建议的方法还避免了单维检验、正态假设、双变量正态性。每种提出的方​​法也避免了主成分分析中用于寻找可靠性 theta 的离群观察和线性假设的问题。几何平均方法为计算 LPI 分数提供了更好的替代方案，避免了权重的缩放和计算满足许多所需的属性，并降低了组件之间的可替代性水平，促进了两个几何平均值的相等性的统计测试，并确定了纠正措施的关键区域。从政策的角度来看，这种识别很重要。一个国家在很长一段时间内的 LPI 图表反映了该国 LPI 的增长模式。如果目标向量被视为表现最佳国家的 LPI 分数，则该方法有助于对国家进行排名和基准测试。JEL 代码：C43、C54 从政策的角度来看，这种识别很重要。一个国家在很长一段时间内的 LPI 图表反映了该国 LPI 的增长模式。如果目标向量被视为表现最佳国家的 LPI 分数，则该方法有助于对国家进行排名和基准测试。JEL 代码：C43、C54 从政策的角度来看，这种识别很重要。一个国家在很长一段时间内的 LPI 图表反映了该国 LPI 的增长模式。如果目标向量被视为表现最佳国家的 LPI 分数，则该方法有助于对国家进行排名和基准测试。JEL 代码：C43、C54